FIG. 1 shows a schematic of a basic linear regulator. The regulator is configured to output a stable output voltage Vout. The output voltage is sensed via a resistive divider (represented by R1 and R2) and fed back to an error amplifier (represented by A). The error amplifier outputs a voltage to the transistor's gate, thereby controlling the amount of current flowing through the transistor and into resistors R1 and R2. The error amplifier is thus able to regulate the output voltage so that the sense-feedback is the same as a reference voltage Vref. The output voltage Vout may therefore be a multiple of the reference voltage Vref.
A linear regulator such as that shown in FIG. 1 is often used to supply a stable voltage to one or more circuits on a chip. These circuits can often require large load currents and good regulation of the output voltage. These requirements increase the current consumption of the regulator, mainly in the error amplifier. Various techniques are available to reduce current consumption in the regulator when the load-current required is lower. One option is to include a load-current dependent bias-component, a technique known as current-boosting. The requirement for a stable system, however, usually limits the current reduction that can be achieved during low-power operation.
Therefore, there is a need for a new voltage regulator suitable for low power implementations.